The Crisis of Keynesian Economics by Geoffrey Pilling (1986)

3 The Foundation of Keynes’ Economics

To historians of economic theory the triumph of the neoclassical synthesis should appear as most inappropriate, for the basis of Keynes’s formal training in the economics of Ricardo and Marshall left a strong imprint on his own contributions to economic theory. It would seem more appropriate to link Keynes’s own theory with the long-period theory of the classical political economists. The possibility of such a relation has become obvious with the post-Keynesian contribution of a long-period theory based on Keynes’s short-period theory which closely resembled, in both content and concern, the classical theory of Ricardo and Marx. (Kregel 1975: xv)[1]

This chapter explores certain aspects of the relationship between the economic theories of Karl Marx and J.M. Keynes from one particular angle, that of the underlying methodologies and general conceptions of these two economic theorists. And because of its importance in current analyses of the Keynesian crisis, we intend to explore these matters from one specific angle: from the standpoint of those who wish to create a new political economy on the basis of a fusion of certain strands within Keynesian economics on the one hand, and some elements from the Marxist-classical tradition on the other. Such efforts involve two distinct, although related questions:

1. That concerning the relationship of the political economy of the classical school to that of Marx. Keynes himself certainly believed that his new economics would undermine what he called the Ricardian foundations of Marxism. In other words, he identified the classical political economy of Ricardo with Marx’s critique of it. We have already commented on this issue and can therefore be brief. Marx’s work involved a critique of political economy, one which understood that there were a series of flaws, fatal in the final analysis, associated with the work of even the best representatives of the school; it was these flaws which made it vulnerable to the attacks of the vulgar, commonplace writers who emerged in the period following Ricardo’s death. (The best treatment of the transition from classical to vulgar economics is provided by Rubin (1979).) The collapse of Ricardian economics was not an event explicable in ideological terms only. That is to say, while Ricardo’s doctrines were certainly attacked because of the subversive uses to which they were being put, not least by the various radical writers in the 1820s and 1830s, the fact remains that the opponents of the classical school did have definite weaknesses in Ricardo’s economics at which to aim their fire and no amount of formal rearranging of the categories of that economics could protect it from its vulgar detractors. Only a fundamental reworking of classical economics which truly transcended it (that is, preserved its positive features while disposing of its negative aspects) could produce a real development of this tradition. It fell to Marx’s lot to make precisely this advance. It is from this standpoint that those attempts made by certain post-Keynesians, as well as by several Marxists, to conflate the work of classical economists and that of Marx are at base erroneous and must be rejected.

2. The second issue implicit in the efforts to reconstitute political economy along lines proposed by Kregel and others involves a certain view of the relationship of Keynes’ work to the classical tradition. In exploring the foundations of Keynes’ economics we shall be concerned with this latter question. Briefly, to anticipate the line of argument, we shall suggest that at a fundamental level these two traditions have little in common and that Keynesian economics is, in the last resort, a continuation, under twentieth-century conditions to be sure, of the vulgar tradition in political economy. We intend to defend this proposition by means of an examination of the categories of The General Theory, suggesting that their subjective and psychological character expresses clearly that they do indeed derive from the vulgar school and not from that represented by Petty, Smith, Ricardo and others.

As we have already seen, the whole of the post-war period has witnessed something of a struggle for the soul of Keynes. On the one hand, in the case of the ‘neoclassical synthesis’ efforts were made to incorporate the teachings of Keynes into the body of neoclassical orthodoxy, producing what Robinson variously castigated as a ‘bastard’ or ‘bowdlerised’ Keynesianism. From a quite different standpoint, equally persistent attempts have been made to bring together certain elements from Keynes’ work with aspects of the classical-Marxian tradition. We shall be concerned not with those who sought to marry the work of Keynes to the prevailing neoclassical orthodoxy but with these latter efforts in order to examine whether, in principle, they are soundly based.

There is no doubt that the emergence of Keynesianism in the 1930s and its later rise to a position of almost unchallenged dominance in the post-war period exercised a decisive influence upon many Marxists operating within the field of economics. Because this is not our prime concern, the historical references will not be explored; but in brief the following can be asserted: under the impact of prevailing Keynesian orthodoxy many Marxists were inclined to read Capital and other of Marx’s economic works from the standpoint of some version or other of an underconsumptionist theory of capitalist breakdown. That is to say they were inclined, as were certain of Keynes’ followers, to see the principal problem for capitalism as lying in a tendency for consumption to fall below the level of that required to sustain investment and full employment, particularly the full employment of labour. The corollary of this position was that appropriate state action, particularly action associated with the state budget, could, by raising the level of consumption, overcome this deficiency.

A book such as Baran and Sweezy’s Monopoly Capital (1966) is a striking case in point. In its overall approach this work is Keynesian, concentrating as it does on what the authors take to be the critical problem for capitalism in the post-war period: the disposal of a rising economic surplus. According to Baran and Sweezy, the big monopoly and near-monopoly concerns are able to fix the prices of their output in such a way as to ensure for themselves ever greater surpluses. This being so, the problem for capitalism boils down to finding ways to absorb this surplus. They see such things as increased advertising expenditures, the economic activities of the state and growing expenditure on arms as the principal means whereby capitalism disposes of its economic surpluses. For them the contradictions of capitalism, especially that between the capitalist class and the working class, no longer exist as they did in Marx’s time. Capitalism is condemned not as an historically limited and inherently contradictory system but as one subjected to increasing irrationality. In this way Baran and Sweezy find redundant the basic categories of Marx’s political economy. Thus of the tendency for capitalism to generate a rising economic surplus they say:

This law immediately invites comparison, as it should, with the classical-Marxian law of the falling tendency of the rate of profit. Without entering into an analysis of the different versions of the latter, we can say that all presuppose a competitive system. By substituting the law of rising surplus for the law of falling profit we are therefore not rejecting or revising a time-honoured theorem of political economy: we are simply taking account of the undoubted fact that the structure of the capitalist economy has undergone a fundamental change since that theorem was formulated. What is most essential about the change from competitive to monopoly capitalism finds its theoretical expression in this substitution.[2]

Their book, published in the middle of the inflationary boom, was an indication of the impact which the prevailing Keynesian orthodoxy had on Marxism.

But the seeds of this move towards Keynesianism had been laid long before – in fact in the 1930s, the decade when Keynes’ major work first appeared. It is noteworthy that Maurice Dobb, for long undoubtedly the leading commentator on Marxist political economy in England, in introducing Sweezy’s earlier and influential exposition of the principles of Marxian economics, The Theory of Capitalist Development, to English readers admits that he had himself moved closer to Sweezy’s heavy emphasis on underconsumptionism as the major factor explaining capitalist crisis, a move which is reflected in much of Dobb’s later writing.

In what is probably still the most satisfactory popular treatment of Marxist political economy, Sweezy had, in this earlier work, when speaking of Keynes as the leading representative of those arguing for liberal capitalist reform, proposed that the critique of such ideas should start ,not from their economic logic but rather from their faulty (usually implicit) assumptions about the relationship, or perhaps one should say lack of relationship, between economics and political action’ (Sweezy 1946: 346). Sweezy is making an important point here, namely, that the question at issue is not so much the economic theory of Keynes but rather the false conception which, as a liberal, he held about the relationship of the economic to the political sphere within the capitalist system. The implication is that Keynesian-type economics was sound in the abstract: the problem arose when one attempted to implement such economics in the ‘real world’ in the face of a state which was not impartial as between social forces an therefore not neutral about policy prescriptions. Keynes is to be rejected not on theoretical grounds, but from the point of view of pragmatism, namely that his ‘solutions’ to the ills of capitalism do not in fact work in practice.[3]

It should be noted in passing that this attitude of Sweezy towards Keynes is remarkably similar to that which Keynes himself took to the work of his predecessors: that it was the faulty assumptions rather than the internal logic of the Manchester School which fatally vitiated its work. Thus, ‘Our criticism of the accepted theory of economies has consisted not so much in finding logical flaws in its analysis as in pointing out that its tacit assumptions are seldom or never satisfied, with the result that it cannot solve the economic problems of the actual world’ (GT: 378).

In our opinion, the position adopted by Sweezy marks a fundamental and wholly unwarranted concession to Keynes. In this chapter we shall suggest that, contrary to Sweezy’s view, the ‘economic logic’ of Keynes was indeed faulty and fatally so; consequently a consideration of this ‘economic logic’ must form the starting point of any sustained Marxist critique of Keynesianism. And this in turn implies that those efforts on the part of ‘left Keynesians’ such as Joan Robinson, to effect some theoretical reconciliation between Marx and Keynes are at base misconceived. The critical distinction drawn by Marx between the classical and the vulgar schools in economics provides a decisive conceptual prism through which to examine certain aspects of Keynes’ economics. For Marx (I, 81) what he dubbed the vulgar school was concerned only with the most superficial aspects of capitalist economy; at a certain point it was a school which degenerated into apologetics, attempting as it did to rationalise away the contradictions of the capitalist system.[4] In particular, Marx castigated the fetishism inherent in vulgar political economy: the tendency to ascribe social powers and functions to things: for instance to attribute the entrepreneur’s ability to make a profit to the natural objects which make up the means of production.[5] We therefore propose to review the main aspects of Keynes’ theoretical system from the point of view of Marx’s distinction between classical and vulgar political economy.

We can start by pointing to certain key features of Keynes’ work which have a direct bearing upon the philosophical and methodological bases of that work. From where, in general, does Keynes begin his analysis? He accepted the fact that capitalism in no way automatically guarantees full employment; there is no self-adjusting mechanism which generates a level of effective demand sufficient to ensure the full utilisation of resources. Keynes’ argument runs along the following well-known lines. In the short run – for him the period of greatest concern, although not necessarily so for many of his post-war followers – the level of employment is a function of the level of output which is in turn a function of the level of effective demand. It is this concept, that of effective demand, which is in reality Keynes’ key idea, a point widely accepted, certainly amongst his orthodox followers (e.g. Patinkin et al.). Effective demand is that demand backed by expenditure. Total expenditure and total sales (assuming no stocks) are the same thing as output. So output is determined by effective demand. Keynes proceeded to break down effective demand into two components – consumption and investment – and analysed each in turn. The basis of the distinction is that money spent on goods and services by individuals to satisfy their own wants is consumption; money spent by enterprises on buildings and machinery in order to produce goods and services in the future is investment. Or, regarding the matter from the standpoint of output, the division is one between investment goods (buildings, machinery, etc.) and consumer goods. To know what determines the level of output – and hence the level of employment – one needs to know what determines the level of consumption and the level of investment. In order to analyse fluctuations in the level of effective demand, Keynes makes use of his three fundamental operational categories – the propensity to consume, the marginal efficiency of capital, and the rate of interest. In combination, these three factors set the limits within which the capitalist economy oscillates, and we shall say more of them presently.

No attempt is made in what follows to provide a full and systematic critique of Keynes’ work. Only a few aspects will be touched upon. It is intended to look briefly at certain of Keynes’ key concepts, to explore their underlying assumptions, concentrating particularly on Keynes’ notion of capital and its ‘productivity’. In view of her central role in the interpretation of both Marx and Keynes, brief comment is made towards the end of the chapter about the work of Joan Robinson. But first the overall approach of Marx to the analysis of capitalism can be sketched out in order to highlight the quite different methodological premises from which his work commences.

For Marx the dynamics of capitalism, the search for the ‘law of motion of modern society’ (‘the ultimate aim of this work is to lay bare the law of motion of modern society’ – Preface: 1) involves as one of its central tasks the investigation of the concept of capital. Despite the claims of empiricism to the contrary, every science has its own hierarchy of concepts; empirical research and individual theories always rest on certain fundamental ideas, forming the cornerstone of the particular department of knowledge concerned. This empiricism denies: it claims to commence from ‘the facts’ unmediated by any preconceptions. This is however pure illusion: all thought begins from definite concepts as to the nature of its object; in this case the capitalist economy. But whether this starting point is a conscious one or not is a matter not without its importance. For those who start with unconscious theoretical categories – that is, from categories which have not been arrived at on the basis of a real critical assimilation of all the past developments in the science concerned – inevitably start their operations from the most vulgar, commonplace definitions. Keynes was here no exception to this law of thought. As we shall see, he did in fact employ an implicit definition of capital and of capitalism, namely as a system of production which was aimed at the satisfaction of human needs. His objection to this system lay in the fact that in its twentieth-century form (where monopoly dominated) it was not doing this as effectively as it could.

There is no doubt that in the case of economic theory the most basic concept with which it deals is that of capital. One of Marx’s most persistent criticisms of the classical school was that it had no real, worked-out, truly consistent concept of capital. Even for Ricardo, the best of the classical economists, capital was merely ‘stored up labour’, a notion which had the effect of making capital coeval with human existence in the sense that even the most primitive tool of the savage represents ‘stored up labour’, the result of past effort on his part, and therefore for Ricardo capital. The seeming universality and general applicability of such an idea was gained, Marx held, at the expense of any real concrete historical content. This is made clear in Marx’s comments on the economists’ conception of capital:

Capital consists of raw materials, instruments of labour and means of subsistence of all kinds, which are utilised in order to produce new raw materials, instruments of labour and new means of subsistence. All these are component parts, are creations of labour, products of labour, accumulated labour. Accumulated labour which serves as a new means of production is capital.

So say the economists.

What is a Negro slave? A man of the black race. The one explanation is as good as the other.

A Negro is a Negro. He only becomes a slave under certain relationships. A cotton spinning machine is a machine for spinning cotton. Only in certain relationships does it become capital. Torn from these relationships it is no more capital than gold is itself money or sugar the price of sugar. (Wage Labour and Capital)

According to Marx, the vulgar economists had an even shallower and fetishistic view of capital. Now the connection between capital and labour in the economics which emerged following the disintegration of the Ricardian school (c. 1830) was entirely lost sight of: the reward to the holders of capital was hereafter held to be a reward for their abstinence or waiting. The unwritten assumption in this view was that capitalism was a system designed to produce wealth for consumption and that those who delayed such consumption had to be appropriately rewarded. Any historical analysis of capital, any inkling that it might express economic relations specific to certain periods and conditions only, fell quite outside the theoretical vision of neoclassical economics.

The Commonplace View of Capital

The outcome of this trend can be seen when we look at the contemporary idea of capital present in orthodox economics. As an example we take the definition of capital offered by the 1984 edition of the Penguin Dictionary of Economics, a summary of similar notions to be found in a hundred textbooks:

The stock of goods which are used in production and which have themselves been produced. ... The word capital in economics generally means real capital – that is physical goods. ... Two important features of capital are (a) that its creation involves a sacrifice, since resources are devoted to making non-consumable capital goods instead of goods for immediate consumption, and (b) that it enhances the productivity of the factors of production, land and labour; it is the enhanced productivity which represents a reward for the sacrifice involved in creating capital. (Bannock et al. 1984: 61)

Two points arise from this definition. (a) Capital is a ‘thing’ merely and not a social relation of production. As such it is presumably coeval with man, indeed perhaps even with the animal kingdom. This was a view which led to a series of strange conclusions and ones which gave Marx much pleasure;[6] (b) despite its seeming lack of social content, this definition does in point of fact carry a very specific conception of the capitalist mode of production. For the reward to capital, the sacrifice of owning it, arises from the fact that immediate consumption must be postponed. Here, implicitly, is the notion that capitalism is a system founded on the satisfaction of human requirements, a system dedicated to providing for the needs of ‘the consumer’ Marx rejected such a petty bourgeois notion because it obscured the fact that under capitalism the aim of production is, and can only be, the self-expansion of capital, that is, its continual accumulation.

Now it is, of course, a truism to say that Keynes criticised certain aspects of the work of the neoclassical school of his day, just as others had done prior to him. But it is equally the case that such criticisms never achieved the rank of anything fundamental, never probed to the epistemological foundations of this school, never inquired into the historical and social conceptions which underlay it. On the contrary, it is apparent that Keynes’ work was itself imbued with precisely the same anti-historical conceptions which predominated in neoclassical economics.

For as is well known, Keynes deliberately abstracts himself from any critical analysis of the social structure of society and its laws of development. In other words, he takes the capitalist system for granted, accepts its appearances as constituting its essence. His concern is exclusively with the functioning rather than the dynamics of capitalism. In his theoretical system he takes both the productive forces and the relations of production to be immutable elements, given once and for all: ‘We take as given the existing skill and quantity of available labour, the existing quantity and quality of equipment, the existing technique, the degree of competition ... as well as the social structure including the forces, other than our variables ... which determine the distribution of the national income’ (GT: 245). Elsewhere Keynes writes that he takes as given (that is, as fixed) the entire economic framework of capitalism (GT: 246).

Now, of course, the fact that Keynes took these factors as ‘given’ does not mean that he was innocent of the fact that, in the empirical sense, this was not the case. A far more serious issue is involved here. It reveals the fact that Keynes’ work involved the conventional and essentially positivist process of model-building whereby, on the basis of a series of arbitrary assumptions, a model of the economy is constructed. That is, Keynes makes a series of assumptions in order to simplify the analysis of the economy – such that there is no technical change taking place, that the ‘economic framework’ of capitalism is fixed – and on the basis of these abstractions a coherent picture of the world is derived. But, as in the case of the traditional assumption of perfect competition, such abstractions are purely mental devices having no basis in the reality of the phenomena being investigated. And precisely because of this they must be arbitrary and subjective. Marx’s analysis is of course based on abstractions (‘In the analysis of economic forms, moreover, neither microscopes nor chemical reagents are of use. The force of abstraction must replace both.’ Preface: 1) but his are abstractions of a quite different order, reflecting as they do the real movement of capital. (For an excellent exposition of Marx’s method of abstraction in Capital and the difference between his procedure and that employed by positivism, see Ilyenkov (1979).)

For Marx all real economic categories – capital, value, rent, interest, profit, etc. – reflect not a series of arbitrary mental assumptions but definite social relations of production. Consequently, they are not categories valid for all epochs and all societies. Let us take the example of capital. According to Marx, capital is no mere thing – raw materials, buildings, factories, etc. – but a social relation which finds expression in or attaches itself to many different things such as money or commodities. The central feature of capitalism, its specifica differentia, the quality which marks it off from past economic systems, is that the means of production are monopolised by a class and face another class, the working class, which is obliged out of necessity to sell its ability to work (in Marx’s terminology, its labour power) to one or other owner of capital. This is why for Marx the essence of capital lies in the fact that it is a social relation and not merely a material thing. Just as the examination of a sack of wheat cannot disclose the social relations under which it was produced (in a feudal demesne, by slave labour, on a collective farm, etc.) so the natural properties of the means of production can never tell us whether they function as capital. A certain class of people may own things such as factories, financial assets and so on but only a definite social process transforms these things into instruments of exploitation, converts them into bearers of that social relationship which Marx designates by the term ‘capital’. Capital is a specific, historically defined, social relation of production. By a fetishistic notion of capital Marx meant precisely that view which tended to ascribe to objects qualities which it was imagined flowed from the material properties of such objects but which in point of fact arose entirely and exclusively from the social role occupied by these things in the process of material production. The notion that things are productive by their nature rather than by virtue of the specific place they occupy in a definite network of social relations was precisely the fetishistic view of capital to which Marx objected.

In its most general form (leaving aside its various types) capital is depicted by Marx in the schema M-C-C’ M’ (M representing the initial sum of money owned by the capitalist). Ignoring here its actual origin, this sum of money is thrown into production, being used initially to purchase commodities, C, including the commodity labour power. In the process of production these commodities are in turn transformed into ones involving a greater amount of (potential) value, C’, which are then sold for an equivalent sum of money, M’. It is from this latter sum of money, M’, that the capitalist meets his individual consumption needs but much more crucially it provides the means for the further accumulation of capital – the reconversion of the surplus value embodied in M’ back into capital.

This schema, M-C-CL-M’, represents in conceptual form the only axis on which production within capitalism takes place. For Marx, capitalism can never be understood if it is seen merely as the production of things satisfying human needs. Were this the case, the limits to production would be purely technical, concerned with the best, most rational, most ‘economic’ allocation of certain material means to the satisfaction of a number of needs. But to envisage the process of production in this way is to ignore the crucial question of the social framework within which production takes place. For capital, its most vital need is to expand M into M’. As Marx says, this is like the law of Moses; should human needs be met in the process of the self-expansion of capital (as of course within limits they are) this is an incidental matter.

This sketch of Marx’s concept of capital reveals, I believe, that he and Keynes approach the analysis of capitalism from quite distinct angles and with fundamentally opposed logics. For although Keynes and Marx both deal in aggregates (in this respect both stand opposed to the old neoclassical approach with its prime focus of interest being the individual), the nature of these aggregates is of a different order. Marx’s principal concern is the total social capital (M-C-M') and its subdivisions; Keynes’ prime interest lies with effective demand and its chief components, investment and consumption.

If the examination of the capitalist process of production proceeds from the standpoint of Marx, it will be immediately evident that the size of the various revenues which in sum constitute the national income depends essentially upon the size of the total social capital and its rate of turnover in the production process. If, for instance, more capital is employed, if more money is transformed from money into commodities needed for production, more will, other things remaining equal, be advanced as variable capital; that is, as wages. In other words, it is the expansion of capital which increases the mass of labour power employed. The empiricist, because he merely records the ‘facts’ has no way of comprehending this process theoretically. The reverse of the real relationship could seem to hold: it could just as well be that if more money is spent on variable capital (wages) more capital will result. Indeed this is exactly how the matter does at first sight appear. ‘It is the absolute movements of the accumulation of capital which are reflected as relative movements of the mass of exploitable labour power, and therefore seem produced by the latter’s own independent movement’ (I: 620). And what is true of wages is true equally of all forms of income – profits, rent, interest, etc. The size of these revenues is limited (determined) by the accumulation of capital, and not the other way round. It is not the size of revenues that fixes the size of the total social capital but the latter which determines the former. To begin one’s analysis with the conditions which determine the turnover of capital is to start from the inner determining source of the movement of the entire capitalist economy. This movement of social capital does of course reveal itself in the size and movement of the various forms of revenue. But to start with these revenues is to commence from a series of immediate, everyday phenomena as they present themselves on the surface of society. And this was just the central point of Marx’s strictures against the school of vulgar economy, namely that it did start uncritically from the immediate economic relations as they appeared on the surface of society. There was no scrutiny of these phenomena to establish their origin, to demonstrate that their roots lay in the specific social relations of a definite economic system, capitalism. The procedure of the vulgar economist did, however, from the point of ideology, carry one advantage: it allowed all the revenues (wages, profits, interest) to be considered of the same status, as ‘factor incomes’ as they say.

To turn specifically to the case of Keynes: a fall in national income will, in the normal run of things, produce a drop in the level of employment; but this merely raises a deeper, more fundamental question: What brings about the initial fall in income? What is the inner (relatively hidden) source of this outer movement? It is this question which, says Marx, any serious analysis of capitalism and its crises must seek to answer. But it is a question which empiricism says does not admit of an answer a search for the inner causes of phenomena is in the last resort deemed futile. According to Marx, the essence of capitalist crises, despite the many differing forms which such crises necessarily take, consists of a break in the conditions of capital turnover, a break in the circuit M-CC’ M’. The movement of capital is the crucial thing from which all else derives and the laws governing the movement of the aggregate capital are the fundamental laws of economics. That is why Marx held that confusion about the nature of capital must lead to confusion about all the other categories of the economy.

Revenue and Capital

The point raised here, the angle from which an analysis of the capitalist economy should begin, is a matter which has continually concerned economics in the past. Let us take the case of Adam Smith. As is widely perceived, Smith held two contradictory theories of value. In places Smith holds to the idea that the value of a commodity is determined by one thing and one thing alone: the quantity of labour embodied in such a commodity. On many occasions, however, he proposed what Dobb and others have appropriately characterised as an adding-up theory of value in which the various forms of revenue (wages, profits, rent) were held, in their summation, to determine the value of the commodity. The conclusion of this latter conception of value is, as Marx puts it, ‘that commodity value is composed of various kinds of revenue, or alternatively “resolved into” these revenues, so that it is not the revenues that consist of commodity value but rather the commodity value that consists of revenues’ (11: 465). Here Marx is raising precisely the same fundamental question as that involved in a critique of Keynes: In the analysis of capitalism, does one begin from value and capital or from price and income (revenue)?

This duality in Smith constituted an unresolved contradiction in his theoretical work: on one hand the effort to discover the inner workings of the economy (here lay the truly classical element in his work), and on the other a mere registering or cataloguing of economic phenomena (here according to Marx are to be found the seeds of the vulgar strand in Smith). Marx attached considerable significance to Smith’s confusion of value and revenue.

But it is this category of ‘revenue’ which is to blame for all the harmful confusion in Adam Smith. The various kinds of revenue form with him the ‘component parts’ of the annually produced, newly created commodity-value, while vice versa, the two parts into which this commodity value resolves itself for the capitalist – the equivalent of his variable capital advanced in the form of money when purchasing labour, and the other portion of the value, the surplus-value, which likewise belongs to him but did not cost him anything – form sources of revenue. (II: 382)

And further:

After starting by correctly defining the component parts of the value of the commodities and the sum of the value-product incorporated in them, and then demonstrating how these component parts form so many different sources of revenue, after thus deriving revenues from value, he proceeds in the opposite direction – and this remains his predominant conception – and turns the revenues into ‘component parts’ into ‘original sources of all exchangeable value,’ thereby throwing the doors wide open to vulgar economy. (II: 372)

Again insisting that we must start from capital if we are to grasp the concept of revenue and determine its size, Marx says:

If I define the length of three straight lines independently and then make these lines ‘components’ of a fourth straight line equal in length to their sum, this is no way the same procedure as if I start with a given straight line and divide this for some purpose or other – ‘resolve’ it so to speak – into three parts. The length of the line in the first case invariably changes with the length of the three lines whose sum it forms; in the latter case the length of the three segments is limited from the beginning by their forming parts of a line of given size. (ibid.: 383)

Keynes’ Key Concepts

We have suggested that the departure points for the theoretical work of Marx and Keynes are of a diametrically opposed character: Marx insists that the crucial starting point for the examination of capitalism is the movement of capital, a movement which alone ultimately explains the nature and size of the various revenues or incomes in capitalist society. Keynes on the other hand starts his analysis from precisely this latter point, from income, or, to use his term, effective demand.

Keynes was concerned with one fundamental problem: the forms which determined the levels of investment and consumption. Now in the first place it is clear that these categories are not in any way unique or specific to capitalism. The consumption of wealth (as food, clothing, shelter) constitutes the material basis for life in all societies, whatever the specific conditions under which that wealth is produced and distributed. Similarly investment – the deployment of a portion of the current social wealth as a means of producing wealth in the future – is by no means an activity unique to capitalism but is present in all economies save the most primitive, where the low level of technique precludes the production of an economic surplus, at least on a regular and systematic basis. What we need to know is the specific form taken by that surplus and the manner in which it is extracted from those who produce it. Keynes here provides no answer for the simple reason that such questions do not enter his horizon.

In this connection, many writers have drawn attention to a striking fact about Keynes’ entire orientation: namely, that the aggregates of his system are not centrally concerned with the specific form taken by consumption and investment within capitalist economy. Thus one writer has said,

One of the significant differences in the methodological character of the aggregate between Marx and Keynes lies in the direction in which abstraction is carried out. Marx’s intention was to represent, as simply as possible, the specific interrelation of aggregates which is characteristic of capitalism, whereas Keynesian aggregates do not necessarily concern themselves with the specifics of capitalism. They are designed primarily to assist in accounting for the total level of employment under the simple assumption that it is proportional to the net national product. (Tsuru 1968)

Marx does not commence his analysis from the standpoint of national income and its division but with the total social capital and its basic disaggregation into constant capital (equivalent to expenditure on machinery, raw materials), variable capital (equivalent to the wages bill) and surplus value (in the form of profit, rent and interest).[7] Not only are these categories specific to capitalism but the contradictions which emerge between them are, according to Marx, an expression of an historically limited mode of production, capitalism. By comparison, the key concepts of The General Theory are abstract in the specific sense that they do not relate to the capitalist economic system as such. Keynes’ theory is based on the proposition that three variables, the propensity to consume (consumption function), the inducement to invest (the marginal efficiency of capital) and the rate of interest (liquidity preference), in their interaction, determine the limits within which the national income fluctuates.

To take first the rate of interest: for Keynes this is determined by the quantity of money and ‘liquidity preference’. (Liquidity preference Keynes defines as the ‘natural’ tendency of people to hold on to liquid assets in the absence of sufficient inducement – in the shape of interest – to relinquish this liquidity.) To understand the nature of liquidity preference it is necessary to know something of the ‘psychological time preference’ inherent in the propensity to consume. According to Keynes, each individual is confronted with two sets of time preferences on which he is obliged to act. First, the individual makes a decision about the proportion of his income to be spent now, and the proportion to be saved for future spending. Having decided the proportion to be saved, he must make a second decision: In what form are these savings to be held? As we know, Keynes proposed that there were three basic reasons for holding money, the speculative motive (the ability to take advantage of anticipated changes in prices) being the decisive one.

The rate of interest is the ‘reward for parting with liquidity for a specified period’, and is determined at the point where the desire to hold a certain amount of cash is just counterbalanced by the pull of the interest rate offered for that quantity of cash. Thus, as Keynes says, the interest rate is a ‘highly psychological phenomenon’ (GT: 202). It is not a payment for waiting or for abstinence as with pre-Keynesian economics, but for not hoarding (GT: 182), for parting with liquidity. Perhaps because Keynes became somewhat obsessed with the parasitic ‘functionless investor’, his theory ignores the fact that interest represents a return on money capital which is of the same general nature and origin as the return on all capital – in short, that interest is a segment of surplus value. Keynes’ theory, which proposes that interest is formed from forces quite independent of the production process, singularly fails to grasp this essential point. This is hardly surprising given that interest-bearing capital, where money seems to breed money, appears prima facie to be quite separate from the production process. As Marx puts it, ‘Capital appears as a mysterious and self-creating source of interest, a thing creating itself. ... The use-value of money ... becomes a faculty of money to generate money and yield interest just as it is the faculty of pear trees to bear pears’ (III: 287).

But economics did not always hold to the sort of fetishistic view of interest as proposed by Keynes. Adam Smith, for instance, says:

It may be laid down as a maxim, that whenever a great deal can be made by the use of money, a great deal will be given for the use of it; and that wherever little can be made by it, less will commonly be given for it. ... The progress of interest, therefore may lead us to form some notion of the progress of profit. (Smith 19176: 105~6)

Interest, Smith implies, is merely part of the profit paid by the industrial capitalist to the money capitalist. Its limits are, therefore, fixed by the magnitude of profit. ‘In any event, the average rate of profit is to be regarded as the ultimate determinant of the maximum rate of interest’ (111: 353). This position of Marx was also held by the best of the classical school for ‘according to the Ricardians and all other economists worth naming, the rate of interest is determined by the rate of profit’ (Th I: 92).

It is for this reason, because profit and interest are both forms of one and the same category, namely surplus value, that they normally move together in the same direction. The demand for liquidity only becomes a potent force in periods of acute economic crisis. The fact that the owners of capital as a whole wish suddenly to transform their capital from its commodity into its money form is itself a graphic expression of a serious rupture in the turnover of capital. According to Keynes’ liquidity preference theory, money assumes the form of a hoard and interest is the reward for not hoarding. In point of fact, however, the function of money as a hoard is but one of its several functions and all of them must be studied in their contradictory unity before we can group the real role of money within the capitalist economy.[8] For instance, one function of money, as everybody recognises, is a means of payment. But this is a contradictory function, a fact by no means universally recognised. For when payments balance each other, money functions only nominally, as money of account, as measure of value. But when actual payments must be made, money no longer acts as a mere intermediary in the process of social metabolism but as the incarnation of wealth in the abstract, as the universal commodity. When, for whatever reason, there is a generalised disturbance in the developed system of payments, money ceases to play its hitherto merely nominal role as unit of account but now becomes the embodiment of social wealth as such. Previously the owners of capital had declared money to be an imaginary creation, and only commodities to constitute real value. Now, in times of sharp crisis, a different cry is heard and ‘as the hart pants after fresh water, so pants the soul after money, the only wealth’ (I: 266). It is thus in times of crises that the demand for money rises sharply and along with it the rate of interest which may now move quite out of line with the rate of profit. Thus, ‘If we observe the cycles in which modern industry moves . . . we shall find that a low rate of interest generally corresponds to periods of prosperity, and a maximum of interest, up to the point of extreme usury corresponds to the period of crisis’ (III: 353). And a little later Marx says,

The rate of interest reaches its peak during crises, when money is borrowed at any cost to meet payments. Since a rise in interest implies a fall in the price of securities, this simultaneously opens a fine opportunity to people with available money-capital, to acquire at ridiculously low prices such interest-bearing securities as must, in the course of things, at least regain their average prices as soon as the rate of interest falls again. (III: 354)

In other words, it is in a crisis, when the rate of profit collapses, that the rate of interest may rise by leaps. It is under conditions of crisis that the rush for liquidity to meet obligations contracted during the phase of prosperity may become a controlling factor. As the pressure for liquidity becomes more generalised, a money famine may occur and bring about a sharp increase in interest rates as the price of other assets falls equally sharply. Many obligations cannot be met and a spate of bankruptcies ensues. It is under these conditions that the demand for liquidity for immediate means of payment becomes so pronounced that to the theoretically untrained eye it may seem to assume an entirely independent existence, such that it can be elevated into the determining cause of the crisis and not seen for what it is, as one of the symptoms of the crisis itself.

The last passage quoted from Marx is interesting in that it suggests that a theory of interest of the type proposed by Keynes took certain phenomena which emerge under conditions of crisis – when, by definition, all monetary and credit relations become violently disrupted – and generalised them into universal principles. As we have already pointed out, for Marx, interest is a return to money capital. It is, from this standpoint, of the same fundamental nature as the return on capital as a whole – it is the payment made out of surplus value earned on the entire capital for the use of a particular portion of that capital. Loan capital depends for its reward on its being successfully employed in the sphere of production. Thus Marx says, ‘Loaning money as capital – its alienation on condition of its being returned after a certain time – presupposes, therefore, that it will be actually employed as capital, and that it actually flows back to its starting point’ (III: 349). One entrepreneur (involved, let us suppose, in vehicle production) shares his total profit with another owner of capital (a banker, let us say) in return for a loan which is to be used with the aim of expanding his capital and surplus value. Returning to the most general form of capital, depicted by the circuit M-C-C'-M', an entrepreneur wishing to expand his capital must be prepared to do several things. First, he must part with his money capital, M. He must turn it into capital having a different form, C, by transforming it into labour power, materials and production equipment. Our entrepreneur must then put these various commodities through a process of production which will turn them into different commodities, C’ which can (hopefully) be sold for an equivalent sum of money, M'. Thus has M been ‘metamorphosised’ as Marx puts it.

Now this initial parting with liquidity occurs quite independently of the proclivities of the owner of capital. Only on condition that capital is initially transformed from M into C can it continue to exist. Marx took great care to stress that in this process capital should not be viewed one-sidedly, as either M or C: it was in point of fact the unity of both these forms, forms which continually passed into each other. Capital was ‘value in motion’ The process of capital accumulation arises not from the inclinations or preferences of the owner of capital: its imperatives stem from the very nature of capital itself and it is precisely because of this that the laws of its accumulation impose themselves upon the individual capitalist, indeed upon the owners of capital as a whole.

On this point, Keynes takes a radically different view. For according to him, the willingness of the owner of money to part with it is at root a psychological matter, not a reflection of the intrinsic needs of capital itself. Here Keynes is quite wrong in that money is always thrown into circulation on definite conditions, ones which are in the final analysis rooted in the realisation of a definite rate of profit. Should such conditions not be met, should there be a sudden collapse in the profit rate, then not only will money capital cease to be committed to circulation but as we have already suggested, the exact opposite can well occur: there will be a general rush for liquidity which will serve both to aggravate the crisis and to force up the rate of interest. (Naturally, miscalculations can be made by the owners of capital as to future profit prospects and under the appropriate circumstances this can clearly aggravate the problems of capital as a whole. But such miscalculations cannot, of themselves, form the basis for an explanation of a crisis.)

In the Keynesian scheme of things the rate of interest and the money supply are abstracted from the process of capital turnover; that is, from the very process in which they alone originate. Keynes’ standpoint is that of the isolated individual who, given certain dispositions, makes a series of decisions about how to hold his wealth. There is here no concrete analysis of how money functions specifically within the capitalist system, and this notwithstanding the fact that Keynes rejected the idea present in some versions of neoclassicism that money was merely a veil, having no independent role. For the money from which the turnover of capital commences in the schema M-C-C'-M' is not money merely but money functioning as capital – in short, money capital. In other words it can be understood only from the point of view of a worked-out scientific conception of capital and it is just this which is lacking in Keynes. Further, the existence of money playing the role not of money but of capital obviously implies the existence of capitalism but also capitalism at a point in its evolution where a differentiation between the various forms of capital has occurred – industrial capital, money capital, commodity capital, etc. If these specific relations are kept in view then the question of liquidity and the rate of interest will be approached in a manner quite different from that of Keynes. Whereas for Keynes the decision whether to hold one’s assets in liquid form depends on expectations about future price movements and the rate of interest, for Marx the investigation of whether the owner of money capital will actually commit such capital to production depends upon the prevailing conditions of production, and critically upon the conditions under which surplus value is being extracted.

A final point in connection with the Keynesian theory of the rate of interest can be made. As we have already noted, for Keynes the rate of interest is determined by the psychology of creditors and by the lending policy of the banking system. To take this second factor, the amount of money in circulation. Here Keynes’ theory is deceptive, confusing as it does the quantity of money and the amount of loan capital. But these two are by no means the same thing (‘the mass of loan capital is quite different from the quantity of circulation’ – III: 499). The former functions as capital, as a necessary initial phase in the circuit of capital. In the event of inflation, given a growth of paper money in excess of the requirements of commodity turnover and prices, an increase in the supply of loan capital which will occur as a result of a growth of temporarily free funds in the bank, will be counteracted by a depreciation of loan capital resulting from inflation. Consequently, in these circumstances there might be no increase in the real supply of loan capital. Here again is revealed the fact that the phenomena of capitalism cannot be judged on the basis of their immediate appearances but require real theoretical analysis if they are to be grasped.

Keynes and the Falling Rate of Profit

In Marx’s opinion, the tendency of the rate of profit to fall – a much discussed and disputed question amongst the classical economists – was the single most important law of political economy. This was so because it was understood that it was the rate of profit which effectively regulated the process of capital accumulation. In other words, profit was important not merely as one of several forms of income within the capitalist system but as the source from which the means for the further accumulation of capital could alone come. In commenting on Ricardo, Marx demonstrates the central importance which he attaches to the profit rate and its tendency to decline, a tendency which Ricardo had himself sensed, if not fully grasped:

The rate of profit is the compelling power of capitalist production, and only such things are produced as yield a profit. Hence the fright of the English economists over the decline in the rate of profit. That the bare possibility of such a thing should worry Ricardo, shows his profound understanding of the conditions of capitalist production ... what worries Ricardo is the fact that the fundamental premise and driving force of accumulation should be endangered by the development of production (III: 254)

The issue is also important in a consideration of the relationship between the economic theory of Keynes and that of Marx in so far as many writers have in the past likened Keynes’ declining marginal efficiency of capital to Marx’s theory of the tendency of the rate of profit to fall.[9]

Before dealing with Keynes’ treatment of this matter, we can briefly set out, in somewhat formal terms, Marx’s general notion of the falling rate of profit. Marx divides the total social capital into three broad categories: (1) constant capital (c), equivalent to expenditure on machinery, raw materials and heat, light and power. This capital was deemed constant in that it merely transfers the value embodied in it and cannot be the source of new value. (2) variable capital (v), the expenditure by capital on the purchase of labour power, variable because it is the only source for the expansion of value. (3) surplus value (s), the increment in value accruing to the owners of capital. The rate of profit is given by surplus value over total capital: s/c + v. Now as capital accumulates, there is a tendency for the constant capital to grow more rapidly than the variable portion of capital: this is the expression in value terms of the improvements in technology associated with capitalism throughout its history. The relatively rapid increase in constant capital as compared with the variable element of capital Marx refers to as the tendency for the organic composition of capital (c/v) to rise. Although an increase in the organic composition of capital will normally produce an increase in the rate of surplus value (s/v), or at least its mass (s), there are definite objective limits to such an increase, not least amongst them the actual physical limit to available working-time. But unless s/v does rise with sufficient rapidity to compensate for the increasing organic composition (c/v), then the tendency for the rate of profit to fall will assert itself in an actual fall.

This is the simplest possible outline of what is in reality a complex law, an outline which ignores both those many counteracting forces to its operation to which Marx drew attention, as well as to the long-standing disputes amongst Marxists about its proper interpretation. But the point to be stressed here is that, as far as Marx is concerned, the tendency for the rate of profit to decline was a product of forces intrinsic to capital. The essential quality of capital is that it is driven to expand and one result of this was the tendency for the rate of profit to fall. Now Keynes, no doubt in an effort to sharpen the impact of his own work, tried to create the impression that all orthodox writers throughout the nineteenth century had subscribed to the notion of a crisis-free capitalism. This was far from being the case; but what did characterise virtually all discussions of the problems associated with capital accumulation was that they were almost invariably seen as being located in disturbances emanating outside the actual social relations of capitalism. (Such was the case with Jevons’ celebrated ‘sunspot’ theory of the trade cycle.)

Bearing this point in mind, we can consider Keynes’ treatment of the movement of the marginal efficiency of capital. By capital Keynes means a thing, a ‘capital asset’ that yields an income. In interpreting capital as a series of assets that produce income, Keynes distinguishes two of its principal forms: ‘instrumental capital’ (a materialised form of capital engaged in the process of production, as in the case of a machine) and ‘consumption capital’ (a material form of capital operating in the sphere of consumption, like, for example, a house) (GT: 226). If we apply this definition to production, it would mean that by capital we have only means of production, that is, employing Marx’s term, constant capital, and not the whole of capital which comprises both constant and variable elements. But in any event, the train of Keynes’ overall argument indicates that by capital he means only its materialised elements, that is the means of production (a view shared by his radical followers, notably Joan Robinson). So it turns out that the marginal efficiency of capital is not to be equated with the Marxian rate of profit, as many appeared in the past to have assumed, since here profit is taken only in connection with constant capital rather than with the whole of capital.

According to Keynes’ theory, any investment which is as yet unutilised will be carried out on one condition, that the anticipated rate of return over the cost of investment exceeds the rate of interest. Given that entrepreneurs aim at profit-maximisation, new investment will be carried on to the point at which the marginal efficiency of capital is equal to the interest rate. The marginal efficiency is determined by two factors: (a) the expected return from an income-yielding asset, and (b) the supply price, or replacement cost of the asset which is the source of that prospective yield. Such a yield takes the form of a flow of income over a period of time, a series of annuities over the anticipated life of the investment; if this stream of yields is then compared with the cost of supplying the assets necessary to produce these yields we have the marginal efficiency of capital, defined by Keynes thus:

More precisely, I define the marginal efficiency of capital as being equal to the rate of discount which would make the present value of the series of annuities given by the returns expected from the capital-asset during its life just equal to the supply price. (GT: 135)

The prospective yields on an asset are undoubtedly for Keynes the key element in determining the marginal efficiency of capital. They are prospective rather than actual because at the moment the investment takes place they are nothing but expectations on the part of the investor. Because of the nature of capital assets, especially those of a long-term durable, large immediate outlays are required before any returns are available to the investor In Keynes’ view capital assets are a link between the known present and an uncertain future.

Now why should the marginal efficiency of capital decline in the long run? Because, ran Keynes’ argument, as more capital is invested it becomes more abundant (less scarce) and therefore produces a lower yield. Subscribing to the scarcity theory of capital, Keynes argues that the returns from capital assets exceed their supply price only because they are scarce (Gr: 213). Every increase in investment brings an increase in output which competes with the output of existing capital. The growing volume of output tends to lower prices and hence to lower too the expected yields from future plant capacity. Keynes of course argued that if the interest rate was kept below the (declining) marginal efficiency of capital schedule then investment could continue unchecked. With ongoing investment, capital would at a certain point cease to be scarce and its return, or marginal efficiency, would then be reduced to zero, a prospect which might be realised within the space of one generation, Keynes felt. (Quite how capitalism could function with a zero rate of return to capital can only remain a sheer mystery from the point of view of the Marxist understanding of such a system. Keynes was in effect proposing the existence of capital without the existence of profit, not a very tolerable state of affairs for the owners of capital one might think!)

A familiarity with the history of economic theory reveals that in Keynes’ explanation for a secular decline in the marginal efficiency of capital is to be found more than an echo of Adam Smith’s theory of the falling rate of profit advanced some 150 years earlier. For it was Smith, following David Hume on this point, who sought to explain the decline in the rate of profit as a result of increasing competition amongst capitals consequent upon accumulation. In The Wealth of Nations Smith says:

The increase of stock, which raises wages, tends to lower profit. When the stocks of many rich merchants are turned into the same trade, their mutual competition naturally tends to lower its profit; and when there is a like increase of stock in all the different trades carried on in the same society, the same competition must produce the same effect in them all. (Smith 1976: 105)

And speaking of the decline of profits in the towns, Smith says

The stock accumulated in them comes in time to be so great, that it can no longer be employed with the ancient profit in that species of industry which is peculiar to them. That industry has its limits like every other; and the increase of stock, by increasing the competition, necessarily reduces the profit. (ibid.: 144-5)

Marx rejected such explanations for the decline in the rate of profit which were centred on competition. We can assume that the ‘forces of competition’ fix the rate of profit at 15 per cent, but the question still remains: Why is this figure 15 per cent? Why not 150 per cent? To ascribe the determination of the rate of profit to competition was, Marx held, to indulge in supply and demand explanations which left unexplored those forces which lay behind supply and demand. It led to a circular argument: the ‘forces of competition’ determined the rate of profit while at the same time the intensity of this competition was measured by precisely the self-same rate of profit. Naturally nobody denies the palpable fact of competition between capitals; Marx’s point however was that this competition is merely the external, outer form in which the inner contradictions of capital expressed themselves. Competition is the realm in which the laws of capital are executed, but such laws are not generated in this sphere.

Keynes’ explanation of the decline in the rate of profit, despite the fact that it is decked out in what superficially appears to be a new terminology, in the last resort rested upon the old neoclassical law of diminishing returns. Marx poured scorn on this ‘law’ first formulated in a clear manner by the eighteenth-century economist, Turgot, as being nothing more than a tautology, based on static assumptions, and in this respect little different from the Malthusian ‘theory’ of population which was also implicitly founded on the assumption of a static technology.

Keynes’ theory of the marginal efficiency of capital was a law which purported to deal with a fundamental secular trend in capitalist economy, an economy characterised by just those forces from which the law abstracts, namely a tendency constantly to revolutionise the techniques of production. It is for this reason that Keynes has been attacked on the lack of realism attaching to this aspect of his theory, a criticism by no means confined to those sympathetic to Marxism. As Schumpeter (1952: 283), for instance, rightly pointed out about Keynes’ theoretical devices: ‘All the phenomena incident to the creation and change in this apparatus [industry], that is to say, the phenomena that dominate the capitalist process, are thus excluded from consideration’.

This quite artificial assumption of a given technology has direct consequences for the Keynesian multiplier theory. This theory deals with the fact that consumption grows as a result of the expansion of production. The growth of Department I production (that producing means of production) actually creates an additional demand for machinery, raw materials, etc. as well as for the means of consumption of workers employed in this department. This in turn implies a growth of output in the other department – Department II, producing means of consumption. But the more rapid growth of the production of means of production signifies the growth of the productive power of social labour and is merely another way of expressing the fact that the organic composition of capital (the ratio of constant to variable capital) is increasing. An increase in the organic composition of capital finds expression in the relative, and in certain cases the absolute, decline in the demand for labour power. The accumulation of capital means the expansion of production on a new, higher level of technology. And this growth in the relative weight of constant as against variable capital will take place in both departments of the economy. The results of this increase in the organic composition depend upon the precise nature of the rise, but one thing does follow: although the accumulation of capital (for the Keynesian, an increase in investment) may bring about an absolute increase in the numbers employed, it will cause a relative lowering of employment. Consequently there is no precise relationship between an increase in investment and an increase in employment; although it might be possible to obtain such a relationship for previous investment, it is one disrupted every time new investment takes place in so far as such investment almost invariably takes place at a new, higher level of productive technique. So here again Keynesianism abstracts from the features specific to capitalism, namely that the development of the productive forces takes the form of a rising organic composition of capital and that production does, in the final analysis, depend on the level of consumption.

There is one final consequence of Keynes’ assumption of a given technology which is worthy of note: namely, that it cannot offer any adequate explanation for the cyclical movement of the rate of profit. If, as capital accumulates and it becomes ‘less scarce’ such that its marginal efficiency declines, then the rate of profit can move in one direction only: downwards. It implies that the fundamental problem of capitalism is one not of the cyclical alteration of booms and slumps but of stagnation and steady decline, as we have noted, precisely the direction in which a number of Keynes’ followers, such as Alvin Hansen in the United States, did indeed interpret his work. Here again it must be emphasised that Marx’s view was of a quite different order. For despite interpretations to the contrary, there is no element of mechanicalism in his conception of the tendency of the rate of profit to fall. Precisely because this law reflected the clash of both objective and subjective factors (which latter included the strength of the working class, its degree of consciousness, the quality of its leadership, etc.), its empirical unfolding could never be known in advance. Capital makes continual efforts to overcome the effects of the operation of this law, but in so doing only raises ever greater obstacles to the smooth, crisis-free expansion of the productive forces: such is the contradictory nature of the capitalist system: ‘Capitalist production is continually engaged in the task of overcoming these imminent barriers, but it overcomes them only by means which place the same barriers in its way in a more formidable size’ (III: 243).

Keynes on Consumption

We can consider briefly the final component of Keynes’ system, namely consumption. Here again the overall tendency of his work, to abstract from the social relations specific to capitalism, is all too evident. (It is noteworthy that orthodoxy on the whole deals with ‘the consumer’ as one of the central actors, if not the central actor, on the economic stage. Naturally, such a person is an empty abstraction, torn from all social and class relations. The fact that ‘everybody is a consumer’ is a proposition which serves to hide the antagonistic divisions which characterise capitalist society.)

Keynes’ propositions about consumption have already been mentioned, are well known, and thus require no more than a brief outline. First, he held that because of the existence of the marginal propensity to consume – according to Keynes a fundamental psychological law – the gap between income and consumption would grow, and unless this gap was appropriately bridged the level of income would fall below that needed to sustain full employment. Various measures were available to raise the level of income including a certain degree of income redistribution, although Keynes was careful to insist that this should be of only ‘moderately conservative’ proportions.

For Keynes ‘consumption’ is the consumption of all individuals in society, each individual being subject to the basic psychological law which he believed determined the relationship between income and consumption. This is far from being Marx’s approach. He drew a distinction between individual consumption on the one hand and industrial consumption on the other. An analogous distinction is that between buyer and consumer. The buyer for Marx is someone who uses up something for his own needs whereas the act of consumption involves using up something in the process of labour. The purchases carried out by the majority of the population within capitalist economy excludes the greater part of the commodities produced in such an economy. For workers buy no instruments of labour, no raw materials; they buy only articles of subsistence, that is, commodities which enter into individual as opposed to productive’ consumption. Marx explains the significance of this point when he says:

This also shows the ambiguity of the word consumer and how wrong it is to identify it with the word buyer As regards industrial consumption, it is precisely the workers who consume machinery and raw materials, using them up in the labour process. But they do not use them up for themselves and they are therefore not buyers of them. Machinery and raw materials are for them neither use-values nor commodities, but objective conditions of a process of which they themselves are the subjective conditions. (Th II: 518)

Marx is here in effect insisting that it is not possible to deal with the level of consumption in the abstract, a-socially, explaining it by reference to a supposed universal psychological disposition on the part of each separate individual. Within the capitalist system consumption takes place always within definite economic (class) relations and it is only by starting with these quite objective relations that the real nature of consumption and its limits can be analysed. As far as the capitalist system is concerned the essential features of these economic relations, as they affect consumption, are (1) the majority of the producers (the working class) are non-consumers (non-buyers in Marx’ terminology) of the greater part of their products, namely the means of production and raw materials, and (2) the majority of producers can consume the equivalent of their product on one condition: that they create surplus value. In short, the level of consumption of the working class cannot be deemed to be determined by the psychological proclivities of a large number of disparate individuals, but by the amount of variable capital (the equivalent to the wage bill) which in turn depends upon the rate of capital accumulation. If this is so then we have but a further illustration of the fact that it is impossible to understand any aspect of capitalist economy unless one starts from the nature of capital and its turnover.

Once Again on the Nature of Capital: The Case of Joan Robinson

We have deliberately concentrated on the conflicting conceptions of capital in the work of Marx and Keynes. This we have done, given that the problem of the essence of capital affects the innermost nature of the production of wealth in contemporary society. We have suggested that Keynes’ treatment of capital is amongst other things characterised by a desire to divorce it from its real relation to production: ‘It is much preferable to speak of capital as having a yield over the course of its life in excess of its original cost, than as being productive’ (GT: 213). So says Keynes, thus in effect reducing the matter to one of semantics only. His sole concern is not with the source of the ‘yield from capital’ but the grounds on which an asset as capital brings a yield in excess of its ‘supply price’. As we know he finds these grounds in the scarcity of such assets.

This chapter has spoken of the decisive importance for Marxists of a theoretically sound critique of Keynes, given the efforts of a number of writers to construct a bridge between Marx and Keynes (albeit a suitably interpreted Keynes). Joan Robinson has here been the decisive figure for such attempts. It goes without saying that Robinson for long occupied a central position in the economic theory of the twentieth century. A member of the famous ‘Cambridge Circus’ which helped Keynes formulate the ideas which produced The General Theory, she has been a leading defender of the Keynes tradition in the years after his death, against both his ‘friends’ as well as his declared enemies. No Marxist, she has nevertheless always claimed considerable sympathy for the ideas of Marx. Politically, throughout the post-war years she identified herself with a number of radical causes. Given these several facets to her work, her views are of considerable interest in the context of the issues raised in this chapter. She has of course also been in the vanguard of those criticising some of the cherished propositions of neoclassical economics, particularly those which deal with the alleged ‘productivity’ of capital. We have not been concerned here with the nature of these criticisms as such. They stem essentially from the work of Sraffa, first begun in the 1920s. The Cambridge school has persistently drawn attention to those problems which orthodox neoclassical theory has encountered in connection with the theory of capital. They can be summarised as follows. This theory contains no prerequisites for aggregating capital goods, that is for discovering the true basis that unites things to form capital and determine its size. Second, in any theory linking the origin and returns on capital with the matter of time a vicious circle ensues: the size of capital is determined by capitalising future revenues, but to establish this method a rate of interest is required; this is however a magnitude the size of which depends upon the amount of capital. This is the nub of the criticisms of orthodox theory made by those in the Sraffa school. (For Marxists it can be said that these problems are the reflection of a decisive confusion, one involving the lumping together of the socio-historical substance of capital with its material form, and in particular one that entails the confusion of ‘capital’ with ‘means of production’. This confusion persists throughout Keynes’ work.)

Now the question is: Has the Cambridge school really got to the bottom of the issue in their criticisms of the neoclassical notion of capital? Joan Robinson outlines her idea of the problem of the productivity of capital when she says:

Whether we choose to say that capital is productive or that capital is necessary to make labour productive, is not a matter of much importance [here, as with Keynes, the matter is reduced to that of a purely semantic problem]. . . . Indeed, a language that compels us to say that capital (as opposed to the ownership of capital) is not productive rather obscures the issue. It is more cogent to say that capital, and the application of science to industry, are immensely productive, and that the institutions of private property, developing into monopoly, are deleterious because they prevent us from having as much capital, and the kind of capital, that we need. (Robinson 1941: 18)

Now, unless one takes the view that, like value, capital is ‘just a name’ the issue as to whether capital is productive, and if so in what precise sense, is by no means a matter of word definitions, but a central, indeed the central question for economic theory. This is demonstrated by amongst other things the history of economics, which has been forced continually to grapple with the mystery presented by capital.[10]

It follows from the standpoint of Marx’s theory of capitalism that only labour (more precisely abstract labour) creates value. But it is by no means a consequence of this view that in the opinion of Marxism the ‘objective factors of production’ (machines, etc.) are to be denied any form of ‘productivity’, as Robinson suggests (‘thus Marx’s refusal to treat capital as a factor of production seems well founded’ (1975: 19)) Quite the opposite is the case: to the extent that such factors raise the level of labour productivity they most certainly contribute to the production of wealth, that is to use values. The word ‘productivity’ however here carries two distinct meanings. First, it can be used to denote the production of use values; it can also indicate that definite social relations are being produced and reproduced. When Marx stressed that capital is productive he did so from a definite angle: as the predominant social relation of capitalist society. And its productivity is from this standpoint quite specific: it is productive of surplus labour which takes the form of interest, profit, etc. And capital was in a position to extract such labour not at all because it involved waiting, because ‘risk’ was involved, was ‘scarce’, functioned as a ‘means of production’ or furthered the ‘application of science to industry’ (Robinson). Capital is productive precisely because it was an essential historical relation for the extraction of surplus labour and far more ‘productive’ in this sense than was feudalism or any other pre-capitalist form of economy. A steam engine in a mine is productive of use values (or rather the labour materialised in such an engine is productive) but this has nothing whatsoever to do with its being capital. It would be equally productive of wealth were it owned by the workers at the mine rather than the entrepreneur. We are dealing here not at all with a matter of words but with a central question: Do we derive the meaning of the word productivity from the relations of man to nature or from the relations of man to man? In other words is there a distinction to be made between productivity in the abstract and something which is productive specifically for capital? In Marx’s opinion not only is there a distinction here but a profound and ever-deepening contradiction: that which is productive for human beings (particularly the working class and other oppressed people) is increasingly positively unproductive (unprofitable) for capital.[11] (Robinson’s distinction between capital and the ownership of capital confuses precisely the essence of the matter: that capital being a social relation cannot exist apart from definite relationships of ownership. Like those utopian socialists of the nineteenth century criticised by Marx, she wants to get rid of the capitalists while retaining capital. But of course to eliminate the one is necessarily to eliminate the other) When Keynes advanced his scarcity theory of the return to capital he was clearly ‘explaining’ its productivity from the former point of view; that is, from the point of view of the relationship of man to nature: capital is productive because it exists in only limited quantity, just as land yields a rent because it is ‘scarce’. And so with Robinson who conflates capital with ‘the application of science to industry’, a purely natural phenomena. [12]

The Concept of Equilibrium

One persistent theme amongst many of the radical Keynesians is their hostility to the notion of equilibrium. Now one thing Marx and Keynes certainly shared in common was their rejection of Say’s law, the notion that capitalism was an automatically self-equilibriating economic system. But their agreement on this point in fact hides more than it tells us because the grounds on which Marx opposed Say were fundamentally different from those of Keynes. We have suggested that in his characterisation of the development of nineteenth-century economic thought, Keynes attached far too much weight to Say’s law of markets. Indeed, as we have noted, in his redefinition of classical economics, Keynes went so far as to make the acceptance of Say’s law the distinguishing criterion for membership of that school, a solecism which enabled him to include not only Ricardo (for Marx the last of the classical economists) but all those who followed Ricardo down to and including Keynes’ contemporary, Pigou.

There is no doubt that the widespread support given to Say’s contention that disequilibrium within capitalist economy is in principle impossible was a significant expression of the increasingly apologetic nature of nineteenth-century economics, as well as a reflection on the part of sections of the middle class for social peace and stability. But it does not follow from this that the denial of any long-run disequilibrium within bourgeois economy marked the real essence of the vulgar school of which Say is the true father. Nor does it follow either that the rejection of the notion of equilibrium, after the fashion of post-Keynesianism, constitutes a sound basis either for the criticism of neoclassical economics or for the establishment of a theory which grasps the real movement of capitalism. We have argued that the degeneration of classical economics resided not in its acceptance of Say’s law (Ricardo had, after all, accepted Say’s proposition in opposition to Malthus[13]) but arose from a deeper, more universal source: in the conscious removal of a consideration of the social relations of production from the province of economics. It was this turn away from an analysis of the (antagonistic) relations of bourgeois economy, a justification of the capitalist system as one based on a natural and eternal harmony of interests, that transformed the science of political economy into the ideology of the vulgar school. In this respect Keynes was far from justified in lumping together Say and Ricardo on the grounds that they shared a common belief in the inherently stable nature of capitalist production, however convenient this device might have been for the pedagogic purposes of The General Theory.

We have attempted to demonstrate that at the level of his basic categories Keynes adopted the standpoint of the vulgar school which started its analysis not from the objective social relations of capitalist economy but from the immediate reflection of those laws in the consciousness of the participants in bourgeois production. Thus when Marx criticised Say’s assertion that capitalism assured the conditions for equilibrium, and automatically so, Marx pointed out that Say was able to reach this (false) conclusion only by ignoring precisely those features which were specific to capitalism. Concretely, when Say proposed that ‘supply creates its own demand’ (Marx called this ‘childish babbling’ and ‘unworthy’ of Ricardo when he repeated it) he had in fact assumed the conditions not of capitalist production but of elementary barter. ‘The conceptions adopted by Ricardo from the tedious Say, that overproduction is not possible, or at least that no general glut of the market is possible, is based on the proposition that products are exchanged against products’ (Th II: 493).

Marx objected to Say’s proposition – that supply and demand would always exist in a state of equilibrium – because it was an empty tautology, emptied, that is, of any social and historical content. Naturally the categories of supply and demand exist within capitalist economy, just as the categories which are the basis of Keynes’ system (investment, consumption, savings, etc.) certainly exist in an empirical sense. But in order to analyse concretely supply and demand within such an economy one had to understand that the production of wealth takes on a specific social form – the production of commodities for the market, and that the supply and demand for commodities was shaped by the feature which dominates in this economy – its division into the two great classes, one which monopolises the means of production and another dependent entirely on the sale of its labour power. In connection with commodity production, a commodity, as something meeting a specific human need is a use value; but at the same time it has a definite exchange value, signifying the fact that it constituted a proportion of total social labour. Marx objected to Say’s empty proposition because in effect it obscured the contradictory nature of all wealth produced within the capitalist economy.

The gist of Marx’s argument on this point runs as follows. Let us take the case of a manufacturer supplying steel. He supplies in a given period of time an amount of steel of a definite use value; say 10 tons of the metal of a certain quality. At the same time he supplies steel of a specific exchange value, signified by its price, 500. But between these two sides of the commodity there is a profound difference which formal thought obscures. For on the one hand the manufacturer places steel with a definite use value on the market which, because of its physical characteristics, is capable of supplying definite needs. At the same time the exchange value of this steel exists only ideally in the shape of a price for the steel which has still to be realised. The seller of the steel is interested in one thing and one thing alone: the exchange value of his steel. He supplies a use value but he is concerned only with the exchange value he will thereby obtain (in money). It is, of course, quite possible for the exchange value of the steel to be expressed in quite different quantities of the metal and indeed this will be the case when there are changes in the productivity of labour in steel-making. The supply of the use value and the supply of the exchange value to be realised are thus by no means identical, since quite different quantities of use value can be represented in the same quantity of exchange value. And just because the exchange value of the steel supplied, but yet to be realised, and the quantity of steel do not coincide, there can be no grounds, a priori, for assuming that there will be no contradiction between these two polar opposites.

The point here is that Marx did not object to Say because he employed the notion of equilibrium as such but because in the proposition that supply and demand always necessarily balance the specific social relations lying behind these abstractions were not considered and nor therefore was the possibility of a contradiction arising between them. Say, in short, reached his conclusions on the basis of empty, purely formal abstractions.

It must never be forgotten, that in capitalist production what matters is not the immediate use-value but the exchange-value and, in particular, the expansion of surplus-value. This is the driving motive of capitalist production, and it is a pretty conception that – in order to reason away the contradictions of capitalist production – abstracts from its very basis and depicts it as a production aiming at the direct satisfaction of producers. (Th H: 495)

And further:

since the circulation process of capital is not completed in one day but extends over a fairly long period until the capital returns to its original form, since this period coincides with the period within which market-prices equalise with cost prices, the great upheavals and changes take place in the productivity of labour and therefore also in the real value of commodities, it is quite clear, that between the starting-point, the prerequisite capital, and the time of its return at the end of one of these periods, great catastrophes must occur and elements of crisis must have gathered and develop, and these cannot in any way be dismissed by the pitiful proposition that products exchange for products. The comparison of value in one period with the value of the commodities in a later period is no scholastic illusion ... but rather forms the fundamental principle of the circulation process of capital. (Th II: 493)

An analysis of capital must take not only the specific social relations of this mode of production into account but must grasp its movement as a whole – in all its interconnected and contradictory moments. The essence of eclecticism is to take bits and pieces from what is a unified process and combine them into a series of empty abstractions. However flexible this may appear to be, however seemingly ‘undogmatic’ such eclecticism seems to the untrained mind, Marx rightly insists upon a different method: in this instance one that aims to grasp capital as a whole in the course of its real development. And if this is the aim of science it becomes impossible to separate out the moments of equilibrium from those of disequilibrium in any absolute sense; this is so because the conditions for the equilibrium of bourgeois economy grow out of the conditions of its disequilibrium, and vice versa. Here the formal method of economics is quite lost. During the period of boom the vulgar eye is directed exclusively to those indices – production figures, growth of trade, expansion of investment – which mirror only the surface outward forms of the capitalist economy. Such empirical ‘facts’ can be compared in any number of ways, and many economists spend their time doing little but just this. In a slump all such indicators tend to be transformed into their opposite. Again, following the prescription of positivism, the indicators can once more be compared in an effort to explain the transformation. But because bourgeois economics does not penetrate beneath the surface of immediate economic ‘data’ (declaring such efforts to be impossible or to involve ‘metaphysics') this transformation, while it may be recorded empirically, can never be understood theoretically. To understand any phenomena theoretically, scientific concepts are essential. And just because these are lacking in orthodox economics, neither the periods of upswing nor the periods of slump which grow organically out of boom conditions can ever be comprehended.[14]

Marx’s analysis of the production of individual capitals could perhaps give rise to the false impression that the sole object of capitalist production – the creation of value and above all the creation of surplus value – is one in which the role of use values can be left out of account. When Marx comes to study the production and reproduction of social capital, that is capital considered as a whole, this is shown not to be the case, for it transpires that this production of value and surplus value is indeed constrained by a barrier which was not taken into consideration in the earlier analysis, namely the barrier constituted by use value on a social scale. In order to reproduce its capital, society must not only have a total fund of value available but it has to find these values ready to hand in a particular useful form; that is, in definite material shape (as machines, raw materials, means of subsistence, etc.). And all these various things must present themselves in proportions determined by the technical requirements of production, proportions which, because methods of production are undergoing continual change, must alter over time.

At the same time, however, Marx’s basic proposition – that capitalism is a system founded on the production and reproduction of surplus value, and a process in which the satisfaction of human needs is an entirely incidental matter – still holds. That is to say, human needs are only met to the degree that satisfying them is a means to the accumulation of surplus value. It is this ever-present growing and developing contradiction between use value and exchange value which lies at the heart of the contradictions of capitalism.

Marx in no way denies the possibility of a solution to this contradiction. But it is the nature of this solution which must be carefully considered.[15] The ‘Reproduction Schemes’ of vol. II of Capital provide the key to grasping this contradiction and manner in which capitalism deals with it. Marx divides social production into two large departments, that producing means of consumption for both the basic classes (Department II), and that producing means of production out of which the existing stock of capital is replaced and extended (Department I). Marx shows how each department is obliged to work for the other, thus establishing a series of complex reciprocal relationships between them. Each department can replace its necessary elements of production only on condition that it obtains a fraction of these elements from the other department and in a suitable material form. On the other hand, each department only comes into possession of the use values it needs if it obtains them from the other department by means of the exchange of equivalent values. In these schemata of reproduction, Marx aims to establish not only the manner in which all the components of the annual value product of society (c + v + s) mutually replace each other. For he also demonstrates how a proportion of the total surplus value produced can be devoted to the further expansion of capitalist production, which naturally presupposes the regular exchange of these value components and their realisation on the market. In this sense the schemes of reproduction in the Capital are an aspect of Marx’s solution to the realisation problem.

It should be noted that this division of the capitalist economy into two basic sections was for Marx no arbitrary one. The product of Department 1 is, in physical terms, machinery and equipment, materials of various kinds such as fuel and electricity which are consumed productively. The products of Department II (food, clothing, housing) can only be used for personal, non-productive consumption. Marx’s central aim, following the example of Quesnay, was to portray the many individual acts of circulation which appear on the surface of society in their characteristic movement, that is in the light of ‘the circulation between the great functionally determined economic classes of society’ (II: 363). Here Marx’s distinction between productive consumption and personal consumption is of a quite different nature from the consumption/investment distinction of Keynes. Whereas Marx’s analytical separation is in the last resort a reflection of the basic class division of society, that of Keynes is devoid of real social content in that for him the distinction between consumption and investment is confined essentially to the question of time.

As we have already indicated, a number of economists – James Mill, Ricardo and Say amongst them – in fact ‘solved’ this problem of the relationship of production to consumption, but only in a superficial manner. They did so by confusing capitalist production (M-C-MI) with simple commodity production (C-M-C) and the latter with barter (C-C). Any act of production, according to this view, creates its own demand, and since, in the last analysis, products are by definition exchanged for products there is an automatic equilibrium of sellers and buyers. But it is a conclusion established not through an investigation of the actual processes of production and circulation in capitalist economy but one arrived at through arbitrary and quite unrealistic assumptions. If the assumptions proposed by Say are accepted the only source of capitalist breakdown will occur if, for whatever reasons, commodities are not produced in the right proportions. In other words, the crisis of capitalism would be one of disproportionality.

Sismondi took a diametrically opposed position on the question of equilibrium within the capitalist economy. Unlike the English classical economists he regarded the commodities appearing on the market not merely as the products of labour but as the products of capital. He believed that capital is able to generate an increase in value; that is, create the conditions for its own self-expansion, because the owner of capital does not pay the full production costs and essentially because he gives the worker an insufficient wage in return for his labour. For him it is precisely this increase in value which provides the source for the accumulation of capital. But then the question must arise: How can the surplus product be sold if the worker who has produced it cannot buy back the portion of the product corresponding to his labour, and if the capitalists themselves do not consume this surplus product (a proposition which follows if a part of it has been capitalised)? Sismondi regarded this as creating an insurmountable problem. He believed that in the final analysis the realisation of surplus product was impossible, unless, that is, it was disposed of, and thus realised, abroad.

Marx rejected both Malthus’ and Sismondi’s ‘solutions’ as being equally one-sided and therefore ultimately false. He did not wish to deny that the realisation of surplus value was a real problem for the capitalist economy. He did however reject Sismondi’s doubts as to the possibility of realisation under capitalism. According to Marx, capitalist production does in fact create its own market and in this way it is able to ‘solve’ the problem of the realisation of surplus value. But it does so not in a metaphysical manner (by abolishing the problem) but in a truly dialectical sense. It solves it, that is to say, by raising this problem to an ever higher and wider – in short, more universal – level. Or, to be more concrete, the realisation problem is resolved only to the extent that the capitalist mode of production advances, only to the extent that it constantly expands its internal and external markets. In this regard, extended reproduction of capital is neither purely impossible, nor can it proceed for ever, uninterruptedly, without breaks, without discontinuities, as the classical economists imagined to be the case. Capitalism ‘solves’ the problem of realisation by taking its internal contradictions to an ever higher level, by continually producing and reproducing them on an ever wider basis until the point where they engulf the whole of society and create the possibilities for the transition to socialism.

For Marx, phases of expansion and of relative equilibrium lead inexorably to their opposites and it is through the form of crises that the contradictions accumulated during the phase of expansion are finally and violently resolved by means of a destruction of capital; this serves to bring the total social capital once more into an appropriate relationship with the total pool of surplus value. Speaking of the many influences which are at work during the period of expansion, Marx says:

These different influences may at one and the same time operate predominantly side by side in space and time and at another succeed each other in time. From time to time the conflict of antagonistic agencies finds vent in a crisis. The crises are always but momentary and forcible solutions of the existing contradictions. They are violent eruptions which for a time restore the disturbed equilibrium. (III: 244)

It is clear that Marx objected to the notion of equilibrium advanced by Say. But one should not conclude from this that such a concept has no place in his analysis. He certainly criticised Say and other economists for having abstracted the notion of equilibrium from the social relations which constitute the real foundation of capitalist economy. This allowed the apologists to make this category absolute. And they were thereby able to declare that equilibrium was a ‘natural” ‘normal’ condition while movement away from it was but a temporary, passing aberration.

But, we stress, this should not lead us to the conclusion that Marx excludes the concept of equilibrium from his work. Let us consider this from the point of view of the law of value. If we assume that two products of labour exchange at their labour value we assume an equilibrium exists between the two given branches of production. Changes in the labour value of a product destroy this equilibrium and cause a transfer of labour from one branch of production to another, thereby bringing about consequent redistribution of the forces of production in the economy. It is changes in the productive powers of labour which cause changes in the amount of labour needed for the production of given commodities, setting in motion corresponding increases or decreases in the value of commodities.

The above mechanism Marx sees as an expression of the functioning of petty commodity production. Under the conditions of capitalist production the process through which equilibrium is simultaneously shattered and established is of a different, higher order. Under capitalism the organisation of production no longer resides in the hands of individual, petty producers, but is now organised by industrial capitalists. Capitalists invest their capital in the sphere of production which is most profitable. The transfer of capital to a given sphere of production creates increased demand for labour power in that particular branch of the economy. As a result, assuming other things to be equal, this brings about an increased price of labour power (wages). This draws living labour into this expanding sector. The distribution of the productive forces amongst the various spheres of the economy and the establishment of conditions of relative equilibrium between them takes the form of the distribution of capitals amongst them. It is the movement of capital, the decisive element of bourgeois production, which is the source of stabilisation and destabilisation alike. Wage labour must move in response to the needs of capital. In short, ‘Wage labour subordinated by capital ... must submit to being transferred in accordance with the requirements of capital and to be transferred from one sphere of production to another’ (III). Specifically it is the movement of capital from those areas with low profit rates to those with higher rates which brings about a tendency towards equilibrium and the establishment of a general rate of profit.

But again this tendency must be seen as a process which is realised not in any mechanical manner. Like all laws it can never appear in unadulterated form. That is, it never produces a situation in which there is actually a general rate of profit throughout the economy. As in the case of all laws, we are faced with a tendency, a contradictory movement towards a never attainable equilibrium, and a movement which must take the form of necessary and constant disruptions. It is this contradictory movement Marx is speaking of when he refers to the ‘incessant equilibrium of constant divergences’ (III: 192).

Now the neo-Keynesians rightly object to the fact that orthodox neoclassical theory treats capitalist economy as though it were a machine, tending by its very nature towards equilibrium. This does not mean however that the whole notion of equilibrium should be consigned to the rubbish bin, as Kaldor and others tend to suggest.[16] The fact is that there is a certain tendency within capitalism towards equilibrium but this is a tendency not to be treated as absolute, nor as a state in which capitalism ‘naturally’ exists but as one attained through just that incessant equilibrium of constant divergences of which Marx speaks. If we consider the movement of the rate of profit and, we repeat, it is this movement which is most significant in the functioning and development of bourgeois economy, it is the existence of the tendency for the establishment of a general rate of profit which serves to pull supply and demand into balance, only to disrupt that equilibrium in the very course of its establishment. Those who see in capitalism only an equilibriating mechanism do of course take a one-sided (and usually an apologetic) view of the capitalist system. But equally one-sided are those who see capitalism as one based exclusively on disequilibrium.

The necessary laws of any series of phenomena find their way, establish themselves, through a maze of deviations. Superficially, such deviations appear as contingent occurrences; yet it is only though such apparent accidents that the law, necessity, establishes itself. At the same time, because it is in such deviations that the tendency is expressed, there are also introduced into the process many new aspects which do not flow from necessity but are conditioned by external circumstances. Take as an instance the Marxist law of value. This holds that there is a necessary relationship between the prices of commodities and their values – the amount of socially necessary labour for its production. This connection manifests itself, and can only manifest itself, in the shape of constant divergences of value from price, first in one direction and then another. Such deviations are, as already indicated, precisely the mechanism through which the general rate of profit is established under conditions where the organic composition of capital – the ratio of constant to variable capital – is not uniform throughout the various branches of the economy.

To underscore the point that Marx recognised that the concept of equilibrium was a necessary abstraction – that is to say, a necessary moment in the real path of capitalist development – we can refer to a passage in which he discusses the function of the general rate of profit. He starts by saying that if we assume that the forces of supply and demand are in equilibrium, we have then to explain the phenomena in which we are interested (in this case price) by means of forces other than those of supply and demand:

If supply equals demand, they cease to act, and for this very reason commodities are sold at their market-values. Whenever two forces operate equally in opposite directions, they balance one another, exert no outside influences, and any phenomena taking place within these circumstances must be explained by causes other than these two forces. If supply and demand balance one another they cease to explain anything, do not affect market values, and therefore leave us so much in the dark as to why the market value is expressed in just this sum of money and no other. (III: 186)

Marx then proceeds to explain why it is necessary to assume, for the purposes of analysis, that supply and demand coincide, even though in reality this is not the case. Such a procedure was necessary ‘to be able to study phenomena in their fundamental relations, in the form corresponding to their conception, that is to study them independently of the appearances caused by the movement of supply and demand.’ And there is an additional reason. This was to allow thought

to find the actual tendencies of their movements and to some extent to record them. Since the inconsistencies are of an antagonistic nature, and since they continually succeed one another, they balance out one another through their opposing movement and their mutual contradiction. Since, therefore, supply and demand never equal one another in any given case, their differences follow one another in such a way – and the result of a deviation in one direction is that it calls forth a deviation in the opposite direction – that supply and demand are always equated when the whole is viewed over a certain period, but only as an average of past movements, and only as the continuous movement of their contradiction. (III: 186).

In examining Keynes’ basic theoretical conceptions we have argued that far from marking any advance on the work of his classical predecessors they constitute a serious degeneration, for whereas Smith, Ricardo and others set out to establish the objective laws of capitalism, Keynes’ work is deeply imbued with the subjectivism which characterises bourgeois thought as a whole in the twentieth century. In the first place, as we have tried to show, his work was highly eclectic, drawing on elements from the neoclassical school for its explanation of the laws of distribution, yet at the same time calling on Malthus for the explanation of the poverty of the 1930s. It was for this reason, because Keynes’ work resembled a rag-bag, that anybody could dip in and choose what they wanted. This is certainly connected with Keynes’ view of the state as a supra-class institution, a point examined in the previous chapter. The state was an institution to be used to direct the economy according to one’s ideas. But this must leave open the question of precisely which policies are to be pursued. Sismondi and Proudhon employed an analysis not unlike that of Keynes to advocate utopian socialist ideas; Malthus used his underconsumptionism to defend the position of feudalism within a rapidly advancing capitalism; in the twentieth century (under quite different historical conditions when capitalism had ceased to be a force for progress) both Fascism and social democracy have operated economic policies which can claim legitimate parentage in Keynes. That such conflicting ideologies are able to find some degree of support in Keynes’ economic theory is no accident given that (a) it was confined to the sphere of circulation (taking the relations of production as given), and (b) it operated with subjective psychological categories.

Keynes’ three independent variables (GT: 246-7) do not even mention profit which for Keynes took back-seat to the gambling instinct which was supposedly inherent in human nature, for ‘If human nature felt no temptation to take a chance, no satisfaction (profit apart) in constructing a factory, a railway, a mine or a farm, there might not be much investment as a result of cold calculation’ (GT: 150). And what are we to make of an economic theory, which after all claimed to explain some of the fundamental problems of twentieth-century capitalism, which could declare: ‘In estimating the prospects of investment, we must have regard, therefore, to the nerves and hysteria and even the digestions and reactions to the weather of those upon whose spontaneous activity it largely depends’ (ibid.: 162)?

Despite its obvious weaknesses, Keynesianism was certainly an important strand in post-war bourgeois ideology. It was the theory which legitimated government spending and the creation of the Welfare State. In the next chapter we shall examine the economic implications of such expenditures.


1. In the same book Kregal makes a similar point when he says: ‘The Keynesian [theory], on the other hand, is more closely linked to Ricardo and Marx of the classical tradition, of the analysis of value in physical terms, the analysis of quantities in terms of some type of measure based on labour, and of the analysis of a system undergoing change through historical time’ (1975: 33).

2. One issue involved in the critique of Baran and Sweezy is their notion of economic surplus. As part of their disposal of the categories of Marx they substitute the notion of economic surplus for that of surplus value. These terms are, of course, by no means the same; all societies, save the most primitive, generate an economic surplus. Only under capitalism does this surplus take the form of surplus value.

3. Keynes was certainly not as accommodating to Marx. He asserted that Marx’s ideas were ‘characterised ... by mere logical fallacy’, and he believed that ‘Marxian Socialism must always remain a portent to the historians of opinion – how a doctrine so illogical and dull can have exercised so powerful and enduring an influence on the minds of men, and through them the events of history’ (Keynes, Laissez-Faire and Communism, quoted in Hunt 1979: 377). Elsewhere Keynes could say, ‘How can I adopt a creed which, preferring the mud to the fish, exalts the boorish proletariat above the bourgeoisie and intelligentsia who, with whatever faults, are the quality in life and carry the seeds of all human advancement?’ (JMK: CEW 9). And this from a man who on the one hand had failed to make even a cursory inspection of Marx’s ideas but nevertheless knew that his own work would destroy its foundations.

4. ‘Once and for all may I state, that by classical Political Economy, I understand that economy which, since the time of W. Petty, has investigated the real relations of production in bourgeois society, in contradistinction to vulgar economy, which deals with appearances only, ruminates without ceasing on materials long since provided by scientific economy, and there seeks plausible explanations of the most obtrusive phenomena, for bourgeois daily use, but for the rest confines itself to systematising in a pedantic way, and proclaiming for everlasting truths, the trite ideas held by the self-complacent bourgeoisie with regard to their own world, to them the best of all possible worlds’ (I: 81).

5. Here empiricism is quite useless as a means to scientific knowledge. Every individual views the world, including its economic phenomena, through social eyes, as an integral part of a definite network of social relations formed historically on the basis of human labour. And because this is so, ‘Socio-historical properties of things very often merge in the eyes of the individual with their natural properties, while transitory properties of things and of man himself begin to seem eternal properties bound up with the very essence of things. These fetishistic naturalistic illusions (commodity fetishism is only one example) and the abstractions expressing them cannot therefore be refuted by mere indication of things given in contemplation’ (Ilyenkov 1982: 127).

6. ‘Political economists have laid it down as an axiom that Capital, the form of property at present predominant, is eternal; they have tasked their brains to show that capital is coeval with the world, and that it has no beginning, so it can have no end. In proof of which astonishing assertion all the manuals of political economy repeat with much complacency the story of the savage who, having in his possession a couple of bows, lends one of them to a brother savage, for a share of the produce of the chase. So great were the zeal and ardour which economists brought to bear on their search for capitalist property in prehistoric times, that they succeeded, in the course of their investigations, in discovering the existence of property outside the human species, to wit amongst the invertebrates: for the ant, in her foresight, is a hoarder of provisions. It is a pity that they should not have gone a step farther, and affirmed that, if the ant lays up stores, she does so with a view to sell the same and realise a profit by the circulation of her capital’ (Lafargue 1975).

7. Strictly speaking variable capital is equivalent to the total wage bill of productively employed workers, that is workers producing surplus value, and not to that of WI workers. So figures such as that for the share of wages in the national income cannot tell us anything directly about such things as the rate of exploitation. The distinction between productive and unproductive labour will be discussed in the next chapter; but there is an indication that the categories of Marx’s Capital do not correspond immediately to empirical data.

8. Economists continually argue amongst themselves about the essential qualities of money. It is conventional to say that money has four functions: (a) as a means of exchange; (b) as a standard of value; (c) as a means of payment; and (d) as a store of value. It was Keynes who laid particular stress on this latter function, making it the basis for his theory of interest: interest was the payment for not hoarding. On the other hand, adherents to the quantity theory of money place their prime emphasis on money’s role as a medium of exchange. Efforts to arrive at the essence of money by means of registering its functions are bound to fail in that they actually stand the real issue on its head. Its functions turn out to be not the manner in which the essential quality of money appears but, on the contrary, the condition from which its nature is deduced. The point is that money manifests several related but contradictory aspects within the capitalist system; to take one aspect, as expressed in one particular function, is bound to lead to an abstract and erroneous conception of money. Thus in the case of mercantilism, an absolute was made of money’s function as a store of wealth and this paved the way for the identification of money with gold and silver. In recent years, at the time of mounting criticism of the position occupied by the dollar in the world monetary system, economists such as Jacques Rueff in France advocated the restoration of the Gold Standard, forgetting that this Standard operated under specific historical conditions during the last century which were incapable of resurrection in the present. On the other hand, those theories which hold that money is purely a convention, employed as a means of fixing relative prices, are equally one-sided. They lead to the conclusion that paper money, rather than the precious metals, are the ideal money-form. Paper money is, however, but one specific form of money and one which arises from its function as means of exchange. The point is that the various functions of money cannot merely be listed but must be considered in their real interconnection.

9. An early example of this was John Strachey (1938), who saw a close analogy between Marx’s theory of the declining rate of profit and Keynes’ notion of the declining marginal efficiency of capital.

10. The mystery of capital consists in the following: How can things (stocks of raw materials, bank balances, machinery and equipment, etc.) so different in appearance be united under the same head as ‘capital’? And second, what is the secret of capital’s ability to expand in value? For an historical account of the various, ultimately futile, efforts on the part of orthodox economics to answer these questions, see Shemyatenkov (1981).

11. At one point Joan Robinson rightly observes that ‘Technical and physical relations, between man and nature, must be distinguished from social relations between man and man’ (Robinson 1960: v). This is indeed the nub of the issue, but it is clear from what she says elsewhere that the real significance of the distinction has eluded her.

12. Joan Robinson suggests that Keynes had a quite new and revolutionary – view of capital: ‘The whole elaborate structure of the metaphysical justification for profit was blown up when he pointed out that capital yields a return not because it is productive but because it is scarce.’ That income arises in connection with a good or service which is naturally or artificially scarce is one of the central features of orthodox rent theory and in this respect Keynes was saying little new. Just as land yields a rent not because it is in scarce supply but because it is privately owned, so the return to various instruments of production reflects not their scarcity but their private ownership as capital. On the similarities between Joan Robinson’s views on capital and those of Proudhon, see Rosdolsky (1977).

13. Given the time that he was engaged in his polemics with Malthus (the early years of the nineteenth century) Ricardo was to an extent justified in assuming that capitalism could develop the productive forces in a smooth crisis-free manner. Such an assumption became far less tenable as the century progressed.

14. The point is that while the world is given to man in sensation it cannot be comprehended through sensation. Empirical material is a necessary component of knowledge and in this sense Marxism is in no way hostile to the study of empirical material; indeed, such study is essential. Marx’s Capital, for instance, involved the exhaustive study of a mass of factual material over a period of some 25 years. But the study of empirical material requires concepts and categories which have to be consciously developed. Those who imagine that they are dealing with ‘the facts’ and the facts alone, invariably operate with the most crass categories of thought which have been uncritically assimilated from bourgeois thought.

15. This involves a conception of the idea of contradiction. In general it can be said that positivism sees in contradiction an error in thought and views the development of thought as always involving the elimination of contradiction. Marxism, on the contrary, sees contradiction as the most vital property of the object itself, and the essential task of scientific thought to be not the elimination of contradiction through the redefinition of terms but as the uncovering of real contradictions and an analysis of their real solution. On this question, see Pilling (1980).

16. Joan Robinson appears not to take this extreme position for ‘The concept of equilibrium, of course, is an indispensable tool of analysis’ (1962: 81).